Khan Academy’s online video tutorials are being hyped to the skies, writes Rick Hess.

Khan Academy isn’t over-hyped, argue Bryan Hassel and Emily Ayscue Hassel on *Ed Next*. It’s mis-hyped. Salman Khan’s “short, engaging tutorials in math, science and other subjects” could be transformative with the addition of a key ingredient: excellent, live teachers.

The Hassels suggest letting students spend part of their school time viewing high-quality videos or smart software, which would replace “teachers’ rote lectures and one-size-fits-few whole group learning.” The best teachers would have time to work closely with more students.

Picture this: let’s say one class out of four in a school’s 4

^{th}grade has an excellent math teacher, and she spends half her instructional time on whole-group instruction and half on more dynamic/personalized learning. If Khan takes over the former whole-group time, two 4^{th}grade classes could have that teacher just for personalized/dynamic learning. The effect is a 100% increase in the number of kids who get a top-tier in-person teacher — without reducing personalized instruction time with kids. She’d need a learning lab monitor for Khan time at school and time-saving digital tools to monitor kids’ progress (a la Wireless Generation or others; Khan’s experimenting with this, too). The change would be at least budget-neutral, andthe great teacher could earn more within budget, since lab monitors are not paid as much.

Technology won’t replace good teachers, the Hassels writes. It can extend their reach.

Some propose “flipping” homework with instruction: Students would view the videos at home and work on solving problems in class. Thirty-nine percent of high school students do no homework, the Hassels write. They won’t watch instructional videos either.

As I have posted previously, I think the Khan philosophy can be a useful tool, especially when teaching procedural skills.

but…

“which would replace “teachers’ rote lectures and one-size-fits-few whole group learning.”

The best whole group teachers I have experienced as a student and as teacher observer have been very good because of the way they interacted with the students. They did not just give a rote, students sit and listen lecture. Lots of questioning and adapting to the answers the students give in order to draw out ideas. (I run in the high school realm therefore this is only applicable to high school.) I don’t think a Khan video can replace this.

I like the idea of the Khan videos, but the ones that I have viewed were very boring. If they are boring to me, I can’t imagine that very many kids are going to sit through them.

Excerpts below from my series taking a critical view of Khan Academy: http://bit.ly/khancritic

Ironically, Khan’s TED talk is in stark contrast to two previous TED talks:

* Dan Meyer – Math Curriculum Makeover http://bit.ly/DanMeyerTED

* Sir Ken Robinson – Do Schools Kill Creativity? http://bit.ly/SirKenTED

According to Meyer, today’s math curriculum is teaching students to expect (and excel at) paint-by-numbers classwork, robbing kids of a skill more important than solving problems: formulating them. How does Khan Academy foster problem posing and creativity?

If your philosophy of education is sit-and-get, i.e., teaching is telling and learning is listening, then Khan Academy (and flipping) are more efficient than in-classroom lecturing.

But why lecture at all? TRUE progressive educators, TRUE education visionaries and revolutionaries don’t want to do these things better. We want to DO BETTER THINGS.

Rather than instructing students with Khan’s videos, teachers should be inspiring them to figure things out on their own and learn how to create their own knowledge by working together. For example, instead of relying on lectures and textbooks, Modeling Instruction in Physics emphasizes active student construction of conceptual and mathematical models in an interactive learning community. Students are engaged with simple scenarios to learn to model the physical world. In comparison to traditional instruction, Modeling is extremely effective — under expert modeling instruction high school students average more than two standard deviations higher on a standard instrument for assessing conceptual understanding of physics.

Watch one Modeling class in action: http://bit.ly/ModelingPhysics

In the clip, the teacher says, “I don’t lecture at all. Instead, I create experiences for the students either in the lab or puzzles and problems for them to solve and it’s up to them to try to figure that out.” I’ve often wondered why this type of teaching hasn’t gotten more attention in the media. Maybe because the teacher is using simple things like whiteboards and bowling balls rather than shiny iPads and SmartBoards?

While Khan argues that his videos now eliminate “one-size-fits-all” education, his videos are exactly that. Plus, they don’t use a lot of the multiple representations that are so fundamental to learning. Concept development is minimal, and he unknowingly plays into student misconceptions. His videos do not align with proper Physics Education Research.

Teachers improve via reading up on pedagogy and getting feedback from mentors & students. Where is Sal’s feedback? Where’s the pedagogy? The research that Khan chooses to ignore is summarized in this one book, now available as a free PDF: “How Students Learn: History, Mathematics, and Science in the Classroom” http://www.nap.edu/catalog.php?record_id=10126

Instead of shifting lecture to home, or making better lectures, we should be shifting away from lectures and doing something better.

Can I just argue that while what Sal Khan is doing is inspirational on many levels, it would be far more effective to have real math teachers teaching the topics, rather than someone who has an okay grasp of the misconceptions students have about mathematics?

Frank-

The key weakness to your own endorsement of modeling comes at the end of the news excerpt you linked to… the class is an AP Physics class, meaning that all those students have likely completed standard physics instruction and are likely amongst the top of their class. I have yet to see an effective way to get students to spontaneously derive various physics formulas from experimentation…just as with any instruction at the top of Bloom’s, physics requires a solid foundation of knowledge and application that can only be effectively taught through teacher-led instruction.

That being said, I am soooooo tired of all these infomercial-style endorsements of educational products that promise to solve all the woes of schools, students, and parents. Nor is Khan academy that revolutionary….teachers and other professionals have been creating short educational videos for years, even on the internet. Khan is just the first individual to make an organized attempt to profit off of it.

Khan is just the first individual to make an organized attempt to profit off of it.“profit off of it”? Source?

“profit off of it”? Source?

Ok, I guess it is possible that he has taken a vow of poverty. True, his organization is nonprofit, but plenty of individuals make quite a bit of money working in nonprofits.

There are several problems with Khan’s proposed “flip”:

First, as Parker points out, lectures are not simply unidirectional talk. They often involve dialogue, questions and answers, etc. A video does not replace this very well.

Second, many students work better alone. They don’t necessarily need “help” with the problems–they need the quiet and the focus in order to work through them. They look to their teachers for intereresting and challenging presentations of material.

Third, if the students are working at their own pace, the teacher must be prepared to teach multiple topics per day and will not have much time for any single one.

The videos are a good resource and could supplement classroom instruction. Beyond that, there are many reasons to be wary.

I use Khan Academy videos to supplement or reteach material in my own classes, or sometimes just to free me up for a few moments to do something else. I also link to them, whether or not I use them in class, so absent students have the opportunity to get *some* instruction during their absence.

Khan is a great *supplement* to what I ordinarily do in the classroom.

From a previous post I made in a different thread…

Speaking from a mathematics perspective, Salman Khan has a great method for the kids who work best with his method – it is great for remediation and for students who tend to need math concepts repeated more than once in order for them to “sink in.” It doesn’t work well with GT kids and upper level kids, because those kids have no opportunity to ask the professor questions about topics related to the topic at hand, or topics regarding extensions of the current topic. Some kids, without the interaction of a knowledgeable professor, would become frustrated with the lessons. That’s why education cannot be packaged into a “one size fits all.” It never will…

In addition, we have tried the “Discovery Method” of teaching mathematics at our school, where the students use problem solving to “discover” the concepts they are supposed to learn. The University of Chicago School Mathematics Project tried this fifteen years ago, and in both instances IT DIDN”T WORK. Unless the students have the foundational knowledge, they will not be able to problem solve effectively. Khan’s videos can give the foundational knowledge. A teacher needs to be there (as the article suggests) to guide the problem solving along…

We’ve already had this conversation.

http://www.joannejacobs.com/2011/03/khan-use-video-to-flip-education/

Again: “The Hassels suggest letting students spend part of their school time viewing high-quality videos or smart software, which would replace “teachers’ rote lectures and one-size-fits-few whole group learning.”

Can anyone find me a classroom in the US where teachers rote lecture in K-8 at all? There hasn’t been a classroom that did that in nearly 30 years. Whole group learning? Are you kidding? Who is left doing something other than differentiated instruction?

The problem with “no-size-fits-all-individual learning of differentiated instruction” in a 30 person classroom is that in 45 minutes, each kid gets less than 2 minutes. No one will learn when the total time with teacher is 2 minutes per class per day.

I imagine discussion roughly similar in form to this one took place among sailors when steamships started showing up.

SuperSub said:“The key weakness to your own endorsement of modeling comes at the end of the news excerpt you linked to… the class is an AP Physics class, meaning that all those students have likely completed standard physics instruction and are likely amongst the top of their class.”Yes, the new reel features a second-year AP class. But Matt Greenwolfe (the teacher in the clip) also teaches his first-year physics course the same way. The Modeling curriculum is specifically designed as a first year physics course, because of the 33% of students who take physics in high school, most only take just one year. Modeling has also been successfully used with 9th graders.

“I have yet to see an effective way to get students to spontaneously derive various physics formulas from experimentation”We do it all the time. Here’s a slideshow I use at Parent’s Night showing our first modeling cycle on constant velocity: http://www.box.net/shared/3hc8aj7yh7

Would you expect kids to derive pi for themselves? The quadratic formula? Even the commutative property?

This is the discussion that keeps on coming up more and more. Just some thoughts from someone who uses a flipped class:

1. It isn’t about the videos, like the author says. Yes, right now, I do use instructional videos, and yes, I produce my own because I want to be the one my students see more frequently. But no, it cannot stop at the videos.

2. Using instructional videos is a great stepping stone for someone that wants to move more in the modeling direction. It is very hard to dive in headfirst, especially if you’ve been doing whole-group instruction for multiple years. Videos are a good way to free up the class time to move into a more modeled environment.

3. Stop looking at the videos as the key to the move. They are simply a tool that works well for some subjects. On the other hand, videos are the last tool I would use for others. You need to make professional judgement of the best method to teach your class. What you do shouldn’t be a duplicate of someone else’s work.

There are pros/cons/arguments for any method of teaching. The common thread of all of them, though, is that a quality teacher MUST be present to help challenge and push our learners.

I sure wish somebody like Kahn would do videos for English grammar.

First of all, you should know that I’m the teacher in the video that Frank referenced above, and Frank’s description of my classroom is correct. About 70% of my students are taking their first physics course, and only 30% are the AP students shown in the video. All the classes are taught with modeling instruction. It just happened that the reporter was there at the time I was teaching that one AP class.

SuperSub said: “I have yet to see an effective way to get students to spontaneously derive various physics formulas from experimentation…just as with any instruction at the top of Bloom’s, physics requires a solid foundation of knowledge and application that can only be effectively taught through teacher-led instruction.”

May I suggest that we may be agreeing about the need for teacher guidance here, but disagreeing about the nature of modeling instruction, which is strongly guided by the teacher. My students do not “spontaneously” derive physics formulas from experimentation. Instead, they learn a set of skills that are glossed over or even ignored in most instruction – careful observation, identification of measurable quantities, identification of independent and dependent variables, experimental design by controlling variables, graphing data, and several graphical analysis techniques for extracting equations from those graphs.

Over the course of a year, they become increasingly independent at carrying out this process. At the start, I interact with the lab groups frequently and show them how to do many of these steps. By the end of the year, I am more herding them – directing them back on course if they’ve veered too far away from a productive path.

Mark Hammond described the model development process of the modeling cycle well in this post, making the same point I’m making. It’s a guided process. http://shawncornally.com/wordpress/?p=2087&cpage=1#comment-4976

Modeling does put the paradigm lab first in the instructional sequence, and those formulas are not coming from anywhere else but the student’s own labs. The results gain an importance and significance that they lack when the labs come last and just confirm received knowledge.

The second half of the modeling cycle is model deployment, in which the model – obtained from the lab – is applied in different contexts and in most cases further developed or refined. I read the University of Chicago “discovery method” study results referred to by Jill Bell to imply that the teaching was lacking the deployment half of the learning cycle. After the lab, in which the students do a lot of the lifting to develop the concept in a concrete situation, they need help in abstracting a principle, and practice in applying it. Most will not spontaneously do these things. They need to be guided. However, guiding them through an authentic scientific process and gradually weaning them towards more independence is much more effective than simply handing them the “facts.”

Stuart Buck asks, “Would you expect kids to derive pi for themselves?” No. But here’s an example of the process. Show students circular objects of a variety of sizes. Ask them what measurements they can take to describe the circles, eliciting diameter and circumference among other things. Ask them how to measure these things – perhaps with a string and a meter stick. After measuring the diameter and circumference of many circles and recording the data in a data table, the results will be graphed. Ask students to predict what the graph will look like and what kind of mathematical function would go with that shape of graph. Then send them loose to do this experiment, along with other investigations they may have come up with during the discussion. Look at the final graphs and ask about the value and meaning of the slope. At the end, did the students derive pi for themselves? No. Did many or most gain some valuable understanding of the meaning of pi that they would not get from simply being handed the equation C = pi*d and working a bunch of rote problems? yes. Do they still need some more practice using the result in other contexts before they see it as a valuable abstract idea? yes.

If you then go on to use the same instructional cycle for your quadratic equation or commutative property, will students start to internalize the process itself and become more independent with it? yes.

Sorry this is so long. Just back from vacation, saw myself referred to here and wanted to reply. Lots to say at once.